A domain‐specific high‐level programming model

Abstract

International audienceNowadays, computing hardware continues to move toward more parallelism and more heterogeneity, to obtain more computing power. From personal computers to supercomputers, we can find several levels of parallelism expressed by the interconnections of multi-core and many-core accelerators. On the other hand, computing software needs to adapt to this trend, and programmers can use parallel programming models (PPM) to fulfil this difficult task. There are different PPMs available that are based on tasks, directives, or low level languages or library. These offer higher or lower abstraction levels from the architecture by handling their own syntax. However, to offer an efficient PPM with a greater (additional) high-levelabstraction level while saving on performance, one idea is to restrict this to a specific domain and to adapt it to a family of applications. In the present study, we propose a high-level PPM specific to digital signal processing applications. It is based on data-flow graph models of computation, and a dynamic runtime model of execution (StarPU). We show how the user can easily express this digital signal processing application, and can take advantage of task, data and graph parallelism in the implementation, to enhance the performances of targeted heterogeneous clusters composed of CPUs and different accelerators (e.g., GPU, Xeon Phi

HeteroPar 2014, APCIE 2014, and TASUS 2014 Special Issue

“…We have selected five papers among the 12 submitted to this special issue: A Domain‐Specific High‐Level Programming Model , which present a high‐level parallel programming model, specifically designed to execute digital signal processing applications on accelerators and regular CPUs. In Execution of Compound Multi‐Kernel OpenCL Computations in Multi‐CPU/Multi‐GPU Environments , the authors address the execution of compound, multi‐kernel, Open Computing Language (OpenCL) computations in multi‐CPU/multi‐Graphic Processing Unit (GPU) environments. The Particle Filter Algorithm: Parallel Implementations and Performance Analysis over Android Mobile Devices , which propose an implementation of the particle filter algorithm (an algorithm frequently used in image and video processing, it constitutes the baseline algorithm in many applications: feature tracking, facial recognition, tracking of vehicles in traffic, video compression, etc.) for mobile devices. Network‐Aware Optimization of Communications for Parallel Matrix Multiplication on Hierarchical HPC Platforms addresses the problem of efficient execution of data‐parallel applications on interconnected clusters and present a topology‐aware optimization that improves data partition by taking into account the entire communication flow of the application. In A High Level and Accurate Energy Model of Parallel and Concurrent Workloads , the authors have developed an energy model and a methodology to automatically extract features that characterize the computational environment relying only on a single power meter that measures the energy consumption of the whole system. …”

“…A Domain-Specific High-Level Programming Model [2], which present a high-level parallel programming model, specifically designed to execute digital signal processing applications on accelerators and regular CPUs. In Execution of Compound Multi-Kernel OpenCL Computations in Multi-CPU/Multi-GPU Environments [3], the authors address the execution of compound, multi-kernel, Open Computing Language (OpenCL) computations in multi-CPU/multi-Graphic Processing Unit (GPU) environments.…”